Add example. Encoder closer to boundaries in best.

s2/README.md

@@ -973,6 +973,21 @@ Valid blocks encoded *without* a dictionary can be decoded with any dictionary.
 There are no checks whether the supplied dictionary is the correct for a block.
 Because of this there is no overhead by using a dictionary.
 
+## Example
+
+This is the dictionary content. Elements are separated by `[]`.
+
+Dictionary: `[0x0a][Yesterday 25 bananas were added to Benjamins brown bag]`.
+
+Initial repeat offset is set at 10, which is the letter `2`.
+
+Encoded `[LIT "10"][REPEAT len=10][LIT "hich"][MATCH off=50 len=6][MATCH off=31 len=6][MATCH off=61 len=10]`
+
+Decoded: `[10][ bananas w][hich][ were ][brown ][were added]`
+
+Output: `10 bananas which were brown were added`
+
+
 ## Streams
 
 For streams each block can use the dictionary.

s2/encode_best.go

@@ -128,7 +128,15 @@ func encodeBlockBest(dst, src []byte, dict *Dict) (d int) {
 				}
 				m := match{offset: offset, s: s, length: 4 + offset, rep: rep}
 				s += 4
-				for s <= sLimit {
+				for s < len(src) {
+					if len(src)-s < 8 {
+						if src[s] == src[m.length] {
+							m.length++
+							s++
+							continue
+						}
+						break
+					}
 					if diff := load64(src, s) ^ load64(src, m.length); diff != 0 {
 						m.length += bits.TrailingZeros64(diff) >> 3
 						break
@@ -157,7 +165,15 @@ func encodeBlockBest(dst, src []byte, dict *Dict) (d int) {
 				}
 				m := match{offset: offset, s: s, length: 4 + candidate, rep: rep, dict: true}
 				s += 4
-				for s <= sLimit && m.length < len(dict.dict)-8 {
+				for s <= sLimit && m.length < len(dict.dict) {
+					if len(src)-s < 8 || len(dict.dict)-m.length < 8 {
+						if src[s] == dict.dict[m.length] {
+							m.length++
+							s++
+							continue
+						}
+						break
+					}
 					if diff := load64(src, s) ^ load64(dict.dict, m.length); diff != 0 {
 						m.length += bits.TrailingZeros64(diff) >> 3
 						break